← All episodes

BillionToOne Is Solving One of Biotech’s Hardest Problems

| 3 products mentioned
Watch on YouTube genetic testing biotech cancer detection liquid biopsy prenatal diagnostics startup scaling product-market fit

BillionToOne has cracked one of biotech's hardest problems: detecting rare genetic signals in blood samples by solving a "needle in a haystack" challenge of finding one abnormal base pair among 3 billion. Founded by two PhD students in 2017, the company went from a half lab bench and $300K to processing 600,000+ tests annually and a $4B+ public valuation by building a step-by-step strategy—starting with prenatal genetic testing before moving to cancer detection—that demonstrates how resource constraints can drive innovation and how founders can scale from idea to market dominance in a regulated industry.

Key takeaways
  • Start with the least capital-intensive problem you can solve in your market, achieve commercial traction, then use those resources to tackle harder, larger problems—this three-step approach (prenatal testing → late-stage cancer detection → early-stage cancer screening) allowed BillionToOne to avoid raising $1B+ upfront and build credibility with each win.
  • Solve the sales problem before scaling operations: BillionToOne had one physician using their test 2 months post-launch; the founders responded by hiring 5 additional sales reps in 3 weeks and pivoting to patient-to-physician marketing, which worked because "when we talk with patients, we can convince them" to request the test from their doctors.
  • Recruit interdisciplinary scientists (not just interdisciplinary teams) who can bridge chemistry, biology, and data science within a single person—BillionToOne's key insight that fetal and tumor DNA amplification creates noise that can be mathematically removed required scientists who understood both the lab chemistry and computational solutions simultaneously.
  • Add synthetic DNA markers to blood samples *before* amplification begins so you can measure exactly how much noise the amplification process introduces, then subtract it from the final data using machine learning—this transforms a "difficult biology problem to almost a simple mathematical problem" that competitors couldn't solve.
  • Structure R&D as internal startups: Create small teams (2-3 research associates + a principal investigator) who report directly to founders, own end-to-end product development, and iterate weekly without bureaucratic friction—this "many startups within the larger company" model accelerates product cycles and keeps top talent engaged even after the company goes public.
  • The holy grail in cancer detection is screening healthy populations for early-stage cancer before symptoms appear; this requires solving the same technical problem as detecting microscopic residual disease in post-surgery patients, so BillionToOne structured their roadmap to reach the harder problem only after proving the technology on easier use cases first.

Mentioned (3)

Y Combinator
Y Combinator "I met Ogazan and David way back in 2017 when they applied to YC." ▶ 1:26
PCR "Every molecular diagnostics approach here requires in the lab using a process called PCR to ampli..." ▶ 3:33
Tesla
Tesla "Is this similar to the Tesla super secret plan the three-step plan to go from like the Roadster t..." ▶ 16:51